ane-speech
Speech recognition on the Apple Neural Engine
Local transcription that leaves the GPU free.
Install · Models · Command line · Python · How it works
ane-speech runs open speech-recognition models on the Neural Engine of Apple silicon Macs. Every layer runs there, so transcription stays quick while the GPU is busy with something else: another model, a render, a game. Nothing leaves your Mac.
It grew out of mlx-speech, whose models it converts, and powers the on-device recognition in TNT.
Install
uv tool install ane-speech # the command line
pip install ane-speech # the Python library, command line included
You need a Mac with Apple silicon and a macOS release that includes Core AI; on an older macOS, ane-speech says so when it starts.
Models
| Name | Model | Best for |
|---|---|---|
confucius4_r2t2 (default) |
Confucius4-R2T2 | Live transcription: text appears while you speak |
qwen3_asr_1_7b |
Qwen3-ASR 1.7B | Whole recordings: one accurate pass once you stop |
Both are multilingual, take a hint of names and terms to expect, and match their MLX originals word for word apart from rare near-ties. Each downloads from Hugging Face the first time you use it. A recording or live take holds up to about a minute of speech.
Command line
ane-speech transcribe meeting.wav --checkpoint qwen3_asr_1_7b
ane-speech transcribe meeting.wav --checkpoint qwen3_asr_1_7b --language Chinese --context "Acme, Kubernetes"
ane-speech stream meeting.wav # R2T2, played as if spoken live, text printed as it settles
ane-speech pull qwen3_asr_1_7b # download ahead of time
Python
import ane_speech
asr = ane_speech.load("qwen3_asr_1_7b")
result = asr.generate("meeting.wav") # a file, or 16 kHz mono float32 samples
print(result.language, result.text)
Live, from a microphone:
live = ane_speech.load("confucius4_r2t2")
session = live.stream_session(chunk_ms=480)
for pcm in microphone: # 16 kHz mono float32, any length
print(session.feed(pcm).committed) # final text, which only grows
print(session.finalize().text)
A loaded model runs in a helper process of its own and exits with yours. The method names follow mlx-speech, so code moves between the two.
Disk space
Each model takes space twice, about its own size each time: the download, in the Hugging Face cache, and a copy compiled for your Mac's Neural Engine. The first load compiles it, which takes a while once, and a macOS update compiles it again. To remove both:
rm -rf ~/.cache/huggingface/hub/models--appautomaton--*-ane
rm -rf ~/Library/Caches/coreai-cache/*/ane-speech
How it works
open checkpoint -> trace in MLX -> convert to Core AI -> run on the Neural Engine
Each model is converted once, in Python: its MLX graph is traced, rebuilt for
the Neural Engine in fp16, and checked against the original. From then on a
small Swift runtime does the work, reading only the converted files, with no
PyTorch. The Python package is a thin layer over that runtime, and other apps
can use the runtime the same way through ane-speech serve, which answers
JSON requests on stdin; TNT does.
Building from source
git clone https://github.com/appautomaton/ane-speech.git
cd ane-speech
swift build -c release
uv run ane-speech transcribe speech.wav # models still come from Hugging Face
AGENTS.md covers converting models yourself, the checks behind them, and releases; a coding agent can follow it as written.
License
The code is MIT. The models keep the licenses of their sources: Qwen3-ASR under Apache-2.0, Confucius4-R2T2 under the NetEase Youdao Model Use License Agreement.
Acknowledgements
The Qwen team for Qwen3-ASR, NetEase Youdao for Confucius4-R2T2, and mlx-speech for the MLX implementations this project builds on.
Metadata
Release files for ane-speech 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| ane_speech-0.1.0-py3-none-macosx_26_0_arm64.whl | Python 3 | none | macOS 26.0+ ARM64 | Details for ane_speech-0.1.0-py3-none-macosx_26_0_arm64.whl |
Release files / ane_speech-0.1.0-py3-none-macosx_26_0_arm64.whl
| Download URL | ane_speech-0.1.0-py3-none-macosx_26_0_arm64.whl |
|---|---|
| Size | 258.6 kB |
| Tags | Python 3 macOS 26.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
e8357fcce35db58205493b6d083dbd9d93dfd00787e9c18f5bef614e53569fe4
|
|
BLAKE2b-256 checksum How to use checksums |
92d53447c4c8b130d403792de591b5794f693be8342363da8290d47f1abe2bdd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 12, 2026.
Transparency logSecurity for ane-speech 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
1 file in this release has been uploaded with provenance metadata.
Attestation for ane_speech-0.1.0-py3-none-macosx_26_0_arm64.whl
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 12, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.